Grafana dashboard for Prometheus Blackbox exporter ICMP ping RTT and packet loss visualization

Prometheus Blackbox exporter is configured to send two types of ICMP ping packets:

1. standard 64 bytes ping packets are sent once in 5 seconds;
2. oversize 64 Kbytes ping packets are sent once in 30 seconds.
   Larger packets may help to diagnose problems with networking or with CPU load being too high.

This dashboard was created with Grafana 9.5.2.

Prometheus Blackbox exporter configuration

modules:
  icmp:
    prober: icmp
    timeout: 5s
    icmp:
      preferred_ip_protocol: ip4
      ip_protocol_fallback: false
  icmp_64kb:
    prober: icmp
    timeout: 5s
    icmp:
      preferred_ip_protocol: ip4
      ip_protocol_fallback: false
      payload_size: 64000

Prometheus scrape configuration

- job_name: 'blackbox-icmp-ping'
  scrape_interval: 5s
  metrics_path: /probe
  params:
    module: [icmp]
  static_configs:
    - targets:
      - localhost
      - server1.example.com
      - server2.example.com
      - server3.example.com
  relabel_configs:
    - source_labels: [__address__]
      target_label: __param_target
    - source_labels: [__param_target]
      target_label: instance
    - target_label: __address__
      replacement: 127.0.0.1:9115  # This is your blackbox exporter.

- job_name: 'blackbox-icmp-ping-64kb'
  scrape_interval: 30s
  metrics_path: /probe
  params:
    module: [icmp_64kb]
  static_configs:
    - targets:
      - localhost
      - server1.example.com
      - server2.example.com
      - server3.example.com
  relabel_configs:
    - source_labels: [__address__]
      target_label: __param_target
    - source_labels: [__param_target]
      target_label: instance
    - target_label: __address__
      replacement: 127.0.0.1:9115  # This is your blackbox exporter.